Analyse maisons - Shapash

Interpretation des predictions maisons

Project_Information

Author : VotreNom

Description : Rapport Shapash pour maisons

Project_Name : Analyse randomforest_maison


Model analysis

Model used : RandomForestRegressor

Library : sklearn.ensemble._forest

Library version : 1.5.2

Model parameters :

Parameter key Parameter value
estimator DecisionTreeRegressor()
n_estimators 174
estimator_params ('criterion', 'max_depth', 'min_samples_split', 'min_samples_leaf', 'min_weight_fraction_leaf', 'max_features', 'max_leaf_nodes', 'min_impurity_decrease', 'random_state', 'ccp_alpha', 'monotonic_cst')
bootstrap True
oob_score False
n_jobs None
random_state None
verbose 0
warm_start False
class_weight None
max_samples None
criterion squared_error
max_depth 16
min_samples_split 2
Parameter key Parameter value
min_samples_leaf 2
min_weight_fraction_leaf 0.0
max_features 1.0
max_leaf_nodes None
min_impurity_decrease 0.0
ccp_alpha 0.0
monotonic_cst None
feature_names_in_ ['etage' 'surface' 'surface_terrain' 'nb_pieces' 'balcon' 'eau' 'bain' 'dpeL' 'dpeC' 'mapCoordonneesLatitude' 'mapCoordonneesLongitude' 'annonce_exclusive' 'nb_etages' 'places_parking' 'cave' 'ges_class' 'annee_construction' 'nb_toilettes' 'ascenseur' 'chauffage_energie' 'chauffage_systeme' 'chauffage_mode'...
n_features_in_ 55
_n_samples 11132
n_outputs_ 1
_n_samples_bootstrap 11132
estimator_ DecisionTreeRegressor()
estimators_ [DecisionTreeRegressor(max_depth=16, max_features=1.0, min_samples_leaf=2, random_state=307423091), DecisionTreeRegressor(max_depth=16, max_features=1.0, min_samples_leaf=2, random_state=1606866035), DecisionTreeRegressor(max_depth=16, max_features=1.0, min_samples_leaf=2, random_state=1621316768),...

Dataset analysis

Global analysis

Training dataset Prediction dataset
number of features NaN 55
number of observations NaN 2,783
missing values NaN 0
% missing values NaN 0

Univariate analysis

etage - Categorical

Prediction dataset
distinct values 1
missing values 0

Target analysis

prix_m2_vente - Numeric

Prediction dataset
count 2,783
mean 2,710
std 951
min 150
25% 2,100
50% 2,700
75% 3,250
max 8,190

Multivariate analysis


Model explainability

Note : the explainability graphs were generated using the test set only.

Global feature importance plot

Features contribution plots

etage -


Model performance

Univariate analysis of target variable

prix_m2_vente - Numeric

True values Prediction values
count 2,783 2,783
mean 2,710 2,700
std 951 684
min 150 695
25% 2,100 2,270
50% 2,700 2,680
75% 3,250 3,130
max 8,190 6,510

Metrics

MAE : 421

R2 : 0.607

MSE : 355,000

MAPE : 0.189

MdAE : 309

Explained Variance : 0.607